GENERAL ARTICLES

Is land really a constraint for the utilization of solar energy in India?

H. Mitavachan and J. Srinivasan* This article compares the land use in solar energy technologies with conventional energy sources. This has been done by introducing two parameters called land transformation and land occupation. It has been shown that the land area transformed by solar energy power generation is small compared to hydroelectric power generation, and is comparable with coal and nuclear energy power generation when life-cycle transformations are considered. We estimate that 0.97% of total land area or 3.1% of the total uncultivable land area of India would be required to generate 3400 TWh/yr from solar energy power systems in conjunction with other renewable energy sources.
Keywords: Land occupation, land transformation, land use, photovoltaic, renewable energy, solar energy. the parameters m2/MW and m2/GWh to compare the land transformations associated with different energy sources. The parameter m2/MW accounts for land area required to set up a typical power plant for each of the energy sources considered; it is the ratio of the area occupied by a typical power plant to its capacity of generation (nominal capacity). The parameter m2/GWh accounts for life-cycle land transformations, which include the area that goes into setting up a power plant, fuel mining (coal and nuclear), transportation (coal only) and waste disposal (nuclear only) across the lifetime of the power plant; it is the ratio of life-cycle land area transformed by a typical power plant to its lifetime energy generation. Land occupation is a measure about how a certain energy source affects the land qualitatively. An approximate calculation for the land transformation and occupation associated with each of the energy sources is discussed in the following sections.

IN a recent article, Sukhatme1 has assessed the potential of different renewable energy sources in India. While assessing the potential of solar energy, he has stated, However, it is fairly clear that the real issue is not the availability of solar radiation as much as the availability of open land. This is going to be the real constraint limiting the use of these sources. In this article, we argue that solar energy differs from other energy sources in terms of land use. We first compare the land-use pattern of three primary energy sources: coal, nuclear and hydro with solar energy. Then, we calculate the percentage of Indias land area that would be required to meet the future projected energy demand, based on the present solar energy technology. Finally, an attempt has been made to answer the question: will availability of land become a limiting constraint for solar energy to become a major player in Indias future electricity power generation mix? We have not considered roof-top PV systems in this study as this issue has been addressed recently by Chokshi2. Technically, the land-use type of any energy technology can be accounted for by two terms land transformation and land occupation. Land transformation is the overall land footprint of the technology across its lifetime which includes, but is not limited to, directly transformed land area for setting up the power plant, mining fuel, fuel transportation, waste disposal and provision of space around the plant. In addition, it also accounts for indirect land transformations the land area that goes into upstream processes and secondary land disturbances, i.e. land degradation due to pollutants and effluents from the fuel and material cycles. In our analysis, we have considered only the direct land transformations. We have used
H. Mitavachan and J. Srinivasan are in the Divecha Centre for Climate Change, Indian Institute of Science, Bangalore 560 012, India. *For correspondence. (e-mail: jayes@caos.iisc.ernet.in) CURRENT SCIENCE, VOL. 103, NO. 2, 25 JULY 2012

Land transformation Coal

A typical coal power plant requires 2023 m2 of land area per MW for plant installation3. Coal power plants not only transform land around the power plant, but they also require land for mining coal and transportation of the extracted fuel from the mines to the plant location. Table 1 gives the break-up of land area transformed by a typical coal power plant in USA. We have used the same values in our study as no data are available for India. Apart from these direct land requirements, coal plants also transform and affect the land indirectly; for example, operating coal mines and building infrastructure require additional land during the upstream processes. In addition, there are secondary land disturbances, such as
163

GENERAL ARTICLES
Table 1. Category Mining Surface Underground Power plant operation Rail transportation attributed to coal Break-up of land area transformed by a typical coal power plant in USA3 Area (m2/GWh) Area (m2/MW) Included Not included

400 200 9.1 3080

Only direct land usage 2023

Indirect land use Secondary land disturbances Area that goes into waste storage

Table 2. Category

Land area transformed by a typical nuclear power plant Area (m2/GWh) 30 15 11804725 48 65 29 120 85 16,05723,140 Area (m2/MW) Reference 3 4 4 3 4 3 3 4

Mining uranium Power plant Power plant including buffer zone Fuel disposal Total land area (including buffer zone)

ever, in this article, we take the direct land footprint of solar technologies, both solar thermal and photovoltaic power generation, as 20,234 m2/MW. Then, the corresponding land area required per GWh of life-cycle electricity produced would be 385 m2, assuming a plant load factor of 20% and the lifetime of the plant as 30 years. Figures 1 and 2 compare the land area required to set up a typical power plant (m2/MW) and the area transformed across its life cycle (m2/GWh) for different energy sources.

contamination of water, land acidification and deterioration of forests, caused by pollutants from coal-fuel cycle.

Land occupation
Land occupation is calculated by multiplying the transformed land area with the time taken to recover to its initial state and hence is a measure of how a certain energy source affects the land qualitatively. Therefore, this term accounts for functional degradation of the land as well. Defining a reference recovery state of land is not an easy task as it depends on the type of vegetation and the local environment interrupted. In their seminal study, Fthenakis and Kim3 have calculated a range of land occupation values for different electricity generation technologies based on life-cycle approach for USA. Figure 3 summarizes the land occupations for the four energy sources in m2 yr/GWh from their study. Further, Fthenakis and Kim3 also argue that the land occupation for photovoltaic decreases with increasing the time that a certain land area is used for generating solar energy, which is contrary to the relatively independent land occupation times of nuclear and other fossil fuel-based energy sources. In the present study, we assume that the values suggested by Fthenakis and Kim3 will hold true for the Indian region as well. More detailed information and discussion about land occupation values, including their calculations, can be found in ref. 3.

Nuclear
The direct land footprint of a nuclear power plant includes power plant area, buffer zone, waste disposal area and the land that goes into mining uranium. Table 2 summarizes the values for land area transformed by a nuclear plant from two international studies3,4.

Hydro
The direct land footprint of hydroelectric power plants varies significantly depending on the geographic location and the primary purpose of the plant. At present, there are around 5100 large dams in India. To assess the land area required for power generation from hydroelectric power generation, we have chosen 9 out of 61 dams which were built mainly for power generation, from the National Register of Large Dams5. Table 3 provides the details of these nine dams. Assuming a plant load factor of 37% and an average lifetime of 50 years, the area required per MW and GWh of electricity produced for a typical dam turns out to be 222,698 and 1374 m2 respectively.

Comparative assessment Solar

The literature suggests that the land area required per MW of installed solar power is around 20,234 m2 (approx. 2 ha)6,7. Due to recent advances in solar technologies, especially the efficiency of solar cells, some authors4,8 give land requirements as low as 12,000 m2/MW. How164

As shown in Figures 1 and 2, although the land area required to set up a solar power plant is more compared to a coal plant, the same does not hold true when we compare it with nuclear and hydro power plants. When we take into account the land area transformed across its life cycle, solar power is next to nuclear and is comparable
CURRENT SCIENCE, VOL. 103, NO. 2, 25 JULY 2012

GENERAL ARTICLES
Table 3. Name Srisailam Chamera-I Salal Baglihar Linganamakki Supa Koyna Rihand Totladoh Year 1984 1994 1986 2009 1965 1987 1964 1962 1989 River Krishna Ravi Chenab Chenab Sharavathy Kalinadi Koyna Rihand Pench Details of large dams considered in this analysis5 Reservoir area (sq. km) 616.4 9.5 9.4 8.1 316.6 123 115 468 77.71 Storage capacity (km3) 4.25 0.11 0.15 4.29 3.76 2.64 8.90 1.09 Installed capacity (MW) 1,670 540 690 450 1090 970 1960 300 160

Only dams that were primarily built for hydroelectric power generation are included.

Figure 1. Mean values of land area required to set up a typical power plant for different energy sources.

Figure 2. Mean values of direct land transformation associated with coal, nuclear, hydro and solar energy sources.

with coal. In fact, the land required for a solar power plant is small compared to hydro power. It should also be noted here that the values considered for coal are US average values. As Indian coal is inferior in quality compared to US coal and, around 70% of coal in India is surfacemined9, the life-cycle land use will go up for this energy source if we take these factors into account. Even though nuclear power is effective in terms of land transformation (life-cycle value), it is the highest when it comes to land occupation (Figure 3). In addition, the recent estimation of required land area for Indian nuclear power plants by Chokshi2 is higher than the one we have assumed here. Hence, there is a need for further work in this direction.

How much of Indias land would be required to meet the future electricity demand by solar energy?
The percentage distribution of Indias land area by landuse type according to the Ministry of Agriculture, New Delhi, is given in Table 4. As India is a densely populated country, the agricultural land and forest cover are necesCURRENT SCIENCE, VOL. 103, NO. 2, 25 JULY 2012

sary for food production and maintaining the ecological balance. Hence, it would be judicious to consider only the waste lands for installing the solar electricity generation systems. Table 4 shows that the total area occupied by waste lands and the land not available for cultivation in India is around 951,860 sq. km, i.e. 31.1% of the total land area. As suggested by Sukhatme1, it would be wise for a densely populated country like India to target at a simple lifestyle pattern with an annual per capita electricity consumption of around 2000 kWh, i.e. 3400 TWh per annum for the country as a whole by 2070. Assuming that the installation of a solar power plant, both photovoltaic (PV) and thermal technologies, requires around 2 ha of land area and redoing the calculations by keeping all other assumptions the same as that by Sukhatme1, we determine that 38,813 sq. km of land would be required to meet the projected annual demand of 3400 TWh. That is, 1.3% of the total land area or 4.1% of the total uncultivable land area, excluding forests and net area sown, is enough to meet the projected demand by solar energy alone. This is less than the land that has been covered by permanent pastures and other grazing land (Table 4).
165

GENERAL ARTICLES
Table 4. Land-use type Forests Land not available for cultivation Permanent pastures and other grazing land Land under tree crops (not included in net area sown) Cultivable waste land Fallow land Net area sown Total Distribution of Indias geographic land on the basis of land-use types10 Percentage of total land 22.8 14.1 3.4 1.1 4.3 8.2 46.1 100 Land area (sq. km) 696,260 432,180 103,880 33,110 131,210 251,480 1,408,610 3,056,740

Source: Directorate of Economics and Statistics, Ministry of Agriculture, New Delhi.

using the projections for other renewable energies by Sukhatme1. Hence, we estimate that 0.97% of the total land area of India or 3.1% of the total uncultivable land area would be required for solar energy to meet Indias future electricity needs in conjunction with other renewable energy sources. It should be noted here that these calculations do not include higher efficiencies achieved by new solar cells. Recently documented efficiencies for different types of solar cells and modules can be found in Green et al.11. The role of concentrated PV technology, which has undergone tremendous development in recent years with its cell efficiencies reaching 40% and higher, is completely ignored in these calculations (see ref. 12 for more information on these technological developments).
Figure 3. Mean values of land occupation for coal, nuclear, hydro and solar energy sources to produce 1 GWh of electricity. Land occupation involves the duration over which the transformed land area returns to its original state, measured as a product of land area and time3.

Discussion
Although we have not accounted for all the land transformations associated with an energy source in our analysis, it is obvious that considering only the land area required to set up the power plant will not give the real picture and hence such an analysis may lead to misconceptions about an energy resource. For example, the solar power plant, which requires more land area (m2/MW) than a coal power plant, transforms less land area than coal power plant with surface-mining option when compared on the basis of life-cycle direct land transformations and is effective in terms of land occupation. It should also be noted that the solar technologies use land statically compared to coal and nuclear energy sources. That is, apparently there is no need for further extraction of resources once a solar power plant has been set up, whereas nuclear and fossil fuel-based technologies must continuously transform some land to extract the fuels or dispose the hazardous waste. Figure 4 shows the water consumption of PV technology in comparison with other technologies13. As PV technology requires minimal amount of water for operation, it provides an opportunity to make use of dry/ waste lands to generate power, which otherwise would be left untouched. In addition, PV technology also
CURRENT SCIENCE, VOL. 103, NO. 2, 25 JULY 2012

Figure 4. Water consumption by the four energy sources for their power plant operations13.

If we bring the other potential renewable energy sources of India into picture, then the land area required by solar technologies to meet the projected annual demand collectively would reduce to around 29,783 sq. km,
166

GENERAL ARTICLES

Figure 5.

Historic summary of the best research cell efficiencies for various photovoltaic technologies14.

Figure 6. A concentrator PV power plant installed by Concentrix Solar. This shows that the technology can support multiple land usage. (Credit: Soitec15).

promises the usage of land for multiple purposes, i.e. the same land can be utilized for grazing, power generation and shading purposes.
CURRENT SCIENCE, VOL. 103, NO. 2, 25 JULY 2012

Further, as cell and module efficiencies increase, even less land will be required for solar power plants. Figure 5 summarizes the efficiency learning curves over the past few decades for various PV technologies. The efficiency of the PV cells has increased from 8% in 1976 to around 43% in 2011 (ref. 14). These newly developed highefficiency multi-junction solar cells are being used in conjunction with concentrators because of their high cost. The concentrators necessitate that the whole system should track the sun accurately. So, the concentrator PV systems have to be mounted on the two axis-tracking towers, as shown in Figure 6 (ref. 15). Notably, this also provides more opportunity to make use of the same land for multiple purposes. Moreover, there is a possibility in the near future that one can lease only the land required for the tracking tower instead of acquiring a large amount of land from farmers (Figure 6). Lastly, land issues are multi-dimensional in nature, which involve societal, political and economical aspects associated with a particular developmental project or energy source in this context. Hence, we cannot assume that the land issues which are affecting the growth of any other energy source will also affect the solar energy source in a similar way. For example, the World Commission on Dams estimates that dam construction submerged 4.5 m ha of Indian forest land between 1980 and
167

GENERAL ARTICLES
2000, and an average dam displaces 31,340 persons and submerges 8748 ha of land16. Because of these issues, land acquiring may be a difficult proposition for hydro power, but these factors cannot play any major role in solar technologies. Similarly, acquiring land to set up a nuclear power plant may be an issue because of the potential risk involved in the technology. For example, the Chernobyl accident contaminated around 300,000 sq. km of land with radio nucleotides17. The same cannot hold true for solar power.
5. National Register of Large Dams 2009, Dam Safety Organisation, Central Water Commission, India; http://www.cwc.gov.in/ main/webpages/dl_index_new.html 6. Hanga, Q., Juna, Z., Xiaob, Y. and Junkui, C., Prospect of concentrating solar power in China the sustainable future. Renew. Sustain. Energy Rev., 2008, 12, 25052514. 7. Fthenakis, V., Mason, J. E. and Zweibel, K., The technical, geographical, and economic feasibility for solar energy to supply the energy needs of the US. Energy Policy, 2009, 37, 387399. 8. Mitavachan, H., Gokhale, A. and Srinivasan, J., A case study of 3-MW scale grid-connected solar photovoltaic power plant at Kolar, Karnataka. Renewable Energy Systems Report, IISc-DCCC 11 RE 1, August 2011; http://www.dccc.iisc.ernet.in/techreports.html 9. Ghose, M. K., Opencast coal mining in India: analyzing and addressing the air environmental impacts. Environ. Qual. Manage., DOI:10.1002/tqem/Spring 2007/71. 10. Table 8.1 Pattern of land utilisation. Ministry of Statistics and Programme Implementation, GoI; http://mospi.nic.in/mospi_new/ site/India_Statistics.aspx?status=1&menu_id=14 11. Green, M. A., Emery, K., Hishikawa, Y., Warta, W. and Dunlop, E. D., Solar cell efficiency tables (Version 38). Prog. Photovoltaics: Res. Appl., 2011, 19, 565572. 12. Kurtz, S., Opportunities and challenges for development of a mature concentrating photovoltaic power industry. Technical Report, NREL/TP-520-43208, revised February 2009. 13. Onat, N. and Bayar, H., The sustainability indicators of power production systems. Renew. Sustain. Energy Rev., 2010, 14, 3108 3115. 14. Research cell efficiency records. National Center for Photovoltaics (NCPV), National Renewable Energy Laboratory, USA; http:// www.nrel.gov/ncpv/, accessed on 22 December 2011. 15. http://www.soitec.com/en/technologies/concentrix/sustainability/, accessed on 11 May 2012. 16. Rangachari, R., Sengupta, N., Iyer, R. R., Banerji, P. and Singh, S., Large dams: Indias experience. A WCD case study prepared as an input to the World Commission on Dams, Cape Town, 2000; www.dams.org 17. Barbalace, R. C., Chernobyl nuclear disaster revisited. EnvironmentalChemistry.com, 1999, accessed on 22 December 2011; http://EnvironmentalChemistry.com/yogi/hazmat/articles/chernob yl2.html ACKNOWLEDGEMENT. We thank Dr Dilip Ahuja, National Institute of Advanced Studies, Bangalore and the referee for their valuable comments and suggestions. Received 23 January 2012; revised accepted 15 June 2012

Conclusion
The present study shows that solar power plants require less land in comparison to hydro power plants, and are comparable to other energy sources including nuclear and coal when life-cycle land transformations are considered. In addition, an attempt has been made to show how solar source differs from other energy sources in the way it uses the land. Because of its unique type of land usage, it has been argued that land availability may not be a limiting constraint for the solar source. Moreover, it should also be noted that viability of solar power vis--vis other forms of power depends on the trade-offs between many factors, such as capital cost, cost of generation, carbon footprint, land area, potential risk involved in the technology, environmental friendliness and many others, but not just on one measure.
1. Sukhatme, S. P., Meeting Indias future needs of electricity through renewable energy sources. Curr. Sci., 2011, 101, 624 630. 2. Chokshi, A. H., Nuclear riddles: TINA and NIMBY. Curr. Sci., 2012, 102, 10961098. 3. Fthenakis, V. and Kim, H. C., Land use and electricity generation: A life-cycle analysis. Renew. Sustain. Energy Rev., 2009, 13, 14651474. 4. Jacobson, M. Z., Review of solutions to global warming, air pollution, and energy security. Energy Environ. Sci., 2009, 2, 148 173.

168

CURRENT SCIENCE, VOL. 103, NO. 2, 25 JULY 2012